search

milaboratory / mixcr

MiXCR is an ultimate software platform for analysis of Next-Generation Sequencing (NGS) data for immune profiling.
https://mixcr.com
Other
316 stars 78 forks source link

low alignment TCR/BCR percentage in whole genome whole blood #631

Closed zuta-osa77 closed 2 years ago

zuta-osa77 commented 3 years ago

Hi,

I have a matched RNA-seq / WGS - seq whole-blood dataset with other data indicating high lymphocyte burden. I used mixcr analyze shotgun as recommended with both datasets.

I have a general question: while the RNA-seq mixcr run for a given sample yields a relatively strong repertoire of BCR/TCR (top clones supported by tens of thousands of reads), the corresponding WGS-seq mixcr run gives a very sparse repertoire (supported by hundreds of reads). Apart from common-sense checks on adapter contamination, which appears to not be the issue, is there something else I might be forgetting as a basic sanity check? Shouldn't I expect a similar sensitivity, indeed an even higher one in the case WGS?

Thanks for a great package, zo

PoslavskySV commented 2 years ago

Hi,

if you still have this issue, please contact us at support@milaboratories.com

Best, Stan

lichennan123 commented 1 month ago

Hello Mixcr team,

I am sorry to reopen this issue because I recently had exactly the same issue. I ran mixcr analyze (exome-cdr preset, according to this discussion on my WBC deep-WGS data. I used adaptor-trimmed fastq files (whole-genome coverage at 150x). But from the alignment report none of the alignment succeeded. My script was attached below for your reference. I wonder if you have any suggestions upon this issue? Thank you.

ml mixcr/4.3.2 mixcr -Xmx50g analyze exome-cdr3 -t40 --species hsa R1.fastq.gz R2.fastq.gz WBC

Chennan

mizraelson commented 1 month ago

Hi, Can you share the reports from the MiXCR output. Also, I recommend using the latest version 4.6.

lichennan123 commented 1 month ago

Here is the content of one representative align.report.txt: align.report.txt

I am definitely gonna try the newest version. But would the same built-in preset 'exome-cdr3' still work with that version? I believe I tried that previously and the preset was not recognized.... Thanks!

Chennan

mizraelson commented 1 month ago

For the new version the command would be:

mixcr analyze exome-seq \
    --species hsa \
      input_R1.fastq.gz \
      input_R2.fastq.gz \
      result

According to your report there are 55795 reads that were successfully aligned. A low alignment rate is expected for non enriched datasets. Also, the reads should cover the CDR3 region at least partially to be successfully aligned for this preset.

lichennan123 commented 1 month ago

Got it. Thank you very much for explaining it. I will give it a shot.